Information-Theoretic Minimax Regret Bounds for Reinforcement Learning based on Duality

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bongole, Raghav, Gouverneur, Amaury, Rodríguez-Gálvez, Borja, Oechtering, Tobias J., Skoglund, Mikael
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913557715615744
author Bongole, Raghav
Gouverneur, Amaury
Rodríguez-Gálvez, Borja
Oechtering, Tobias J.
Skoglund, Mikael
author_facet Bongole, Raghav
Gouverneur, Amaury
Rodríguez-Gálvez, Borja
Oechtering, Tobias J.
Skoglund, Mikael
contents We study agents acting in an unknown environment where the agent's goal is to find a robust policy. We consider robust policies as policies that achieve high cumulative rewards for all possible environments. To this end, we consider agents minimizing the maximum regret over different environment parameters, leading to the study of minimax regret. This research focuses on deriving information-theoretic bounds for minimax regret in Markov Decision Processes (MDPs) with a finite time horizon. Building on concepts from supervised learning, such as minimum excess risk (MER) and minimax excess risk, we use recent bounds on the Bayesian regret to derive minimax regret bounds. Specifically, we establish minimax theorems and use bounds on the Bayesian regret to perform minimax regret analysis using these minimax theorems. Our contributions include defining a suitable minimax regret in the context of MDPs, finding information-theoretic bounds for it, and applying these bounds in various scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2410_16013
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Information-Theoretic Minimax Regret Bounds for Reinforcement Learning based on Duality
Bongole, Raghav
Gouverneur, Amaury
Rodríguez-Gálvez, Borja
Oechtering, Tobias J.
Skoglund, Mikael
Machine Learning
Information Theory
We study agents acting in an unknown environment where the agent's goal is to find a robust policy. We consider robust policies as policies that achieve high cumulative rewards for all possible environments. To this end, we consider agents minimizing the maximum regret over different environment parameters, leading to the study of minimax regret. This research focuses on deriving information-theoretic bounds for minimax regret in Markov Decision Processes (MDPs) with a finite time horizon. Building on concepts from supervised learning, such as minimum excess risk (MER) and minimax excess risk, we use recent bounds on the Bayesian regret to derive minimax regret bounds. Specifically, we establish minimax theorems and use bounds on the Bayesian regret to perform minimax regret analysis using these minimax theorems. Our contributions include defining a suitable minimax regret in the context of MDPs, finding information-theoretic bounds for it, and applying these bounds in various scenarios.
title Information-Theoretic Minimax Regret Bounds for Reinforcement Learning based on Duality
topic Machine Learning
Information Theory
url https://arxiv.org/abs/2410.16013